{"repo":"NousResearch/hermes-agent","public_id":"111092","display_label":"Gateway startup watchdog and checkpoint maintenance fixes","summary":"These issues and fixes address the gateway startup watchdog prematurely killing large installs during pre-logging phases and maintenance operations. The work focuses on ensuring progress leases are held across auto-maintenance, checkpoint GC, and state.db operations so the watchdog does not interrupt valid startup or tool calls.","subsystem":null,"counts":{"members":5,"open_issues":0,"open_issues_distinct":0,"open_prs":0,"closed_issues":1,"merged_prs":2,"attempt_prs":2},"landed_fix":{"status":"merged_to_main","basis":"structural","pr":111460},"resolution":{"state":"resolved","by":"pr-111460"},"freshness":{"state":"verified","last_verified_at":"2026-09-15T17:57:17+00:00","is_stale":false},"grouping":{"backbone":"STRUCTURAL_SPANNING","link_density":0.4},"fix_edges":[{"issue_number":79334,"pr_number":112063,"tier":"verified","basis":"superseded_ref","confidence":0.9,"recoverable_from_timeline":false},{"issue_number":111092,"pr_number":111460,"tier":"verified","basis":"closing_ref","confidence":0.95,"recoverable_from_timeline":true},{"issue_number":111092,"pr_number":112063,"tier":"verified","basis":"superseded_ref","confidence":0.9,"recoverable_from_timeline":false}],"members":[{"kind":"issue","number":111092,"title":"[Bug]: Gateway restart live-lock on large installs — pre-logging startup phases hold no progress leases; startup watchdog kills each attempt (exit 75) before the first log line","url":null,"state":"closed"},{"kind":"pr","number":111120,"title":"fix(gateway): hold startup-watchdog progress leases across state.db auto-maintenance","url":null,"state":"closed","close_reason":{"category":"implemented_elsewhere","implication":"superseded"},"merged":false},{"kind":"pr","number":111460,"title":"fix(gateway): startup watchdog no longer kills a large install mid-maintenance and always keeps its stack dump (#111092, salvage #111120 #111124)","url":null,"state":"merged","close_reason":{"category":"merged","implication":"superseded"},"merged":true,"is_fix_pr":true,"fix_tier":"verified","files_changed":["hermes_startup_watchdog.py","hermes_state_maintenance.py","hermes_state_sessions.py","tests/gateway/test_startup_watchdog.py","tests/hermes_state/test_startup_maintenance_lease.py","website/docs/reference/environment-variables.md"]},{"kind":"pr","number":111980,"title":"fix(gateway): checkpoint auto-prune also renews the startup watchdog lease","url":null,"state":"closed","close_reason":{"refs":[{"kind":"commit","value":"804707bea620b","number":0},{"kind":"pr","value":"#111460","number":111460},{"kind":"pr","value":"#112063","number":112063}],"category":"superseded_by_pr","implication":"superseded"},"merged":false},{"kind":"pr","number":112063,"title":"Checkpoint store gc no longer stalls tool calls, gateway startup or hermes update (refs #79334, supersedes #79335, #111980)","url":null,"state":"merged","close_reason":{"category":"merged","implication":"superseded"},"merged":true,"is_fix_pr":true,"fix_tier":"verified","files_changed":["cli.py","gateway/run.py","hermes_cli/config_defaults.py","tests/gateway/test_checkpoint_prune_housekeeping.py","tests/tools/test_checkpoint_manager.py","tools/checkpoint_manager.py","website/docs/user-guide/checkpoints-and-rollback.md"]}],"trust":{"default_tier":"verified","includes_suggested":false}}